Despite the proliferation of formal and informal research mentoring initiatives, relatively little is known about specific factors in mentoring relationships that account for the positive effects on student outcomes. Studies are needed that identify causal mechanisms of research mentoring in order to develop, implement, and evaluate the effectiveness of innovations such as research mentor training interventions aimed at advancing the science pursuits of underrepresented minorities (URM). The proposed research addresses this critical need through the following three Specific Aims: Specific Aim #1. Establish the psychometric properties of the research mentor and mentee surveys currently in use to evaluate the Wisconsin Mentoring Seminar (WMS) and establish evidence for their construct validity and reliability. Approach: Aim #1 will be accomplished using item and exploratory factor analysis to establish correlated variables and reveal underlying elements of research mentoring relationships. Specific Aim #2. Identify critical elements in research mentoring relationships associated with student outcomes. Approach: Aim #2 will be accomplished using hierarchical multiple regression and path analysis to test theoretically-proposed relationships among research mentoring elements and student outcomes. Specific Aim #3. Modify the WMS in response to Specific Aim #2 and test the effectiveness of this adapted research mentor training intervention on student outcomes. Approach: Aim #3 will be accomplished by employing a two-group pretest- posttest design to test the impact of trained research mentors and untrained research mentors on student outcomes. A repeated measures analysis of variance will be performed on the data. This proposal is innovative because it: employs a quasi-experimental research design, providing comparison between trained mentors and untrained mentors; uses retrospective and prospective, longitudinal data; and applies theoretically-grounded and methodologically rigorous approaches. This project is significant because it will yield: 1) psychometrically sound measures of the research mentoring relationship, 2) a tested, theoretically-based research mentor training intervention that promotes science careers for URM students, and 3) immediate translation of findings into directions for future adaptation of research mentor training efforts. PUBLIC HEALTH RELEVANCE: The National Institute of General Medical Sciences prioritizes research training programs to produce the next generation of scientists who can meet the growing health needs for the public. The goal of the proposed work is to advance the representation of traditionally underrepresented groups in biological research by studying the predictors, correlates, and outcomes of research mentoring and a research mentor training intervention aimed at promoting their pursuit of science careers. Thus, the findings are ultimately expected to advance the culturally diverse workforce needed to address health disparities both in disease prevention and medical treatment.